Florida Lake Management Society Annual Conference, Naples, Florida, June 4 – 7, 2007
Preliminary results of the analysis of the 2006 data suggest increased classification accuracy for SAV mapping as a consequence of greater spatial and spectral resolution and an increase in ground-truthing conducted during imagery acquisition. Spectral signatures show significant between-class variation, which should lead to more robust image classification throughout all remaining images. Classified images from both the 2003 and 2006 data collection were also compared to groundtruthing transects that were conducted near the imagery collection date. Transect data were used to assess the accuracy of the supervised classification techniques applied to the remotely sensed data.
A methodology was developed for processing the 2003 airborne hyperspectral imagery and fieldwork reflectance data into a spectral library suitable for SAV identification. This same methodology was improved upon with the addition of higher resolution imagery collected in 2006. Hyperspectral imagery has proven to be very useful for SAV identification with the caveat that water depth and existing conditions (algal blooms, existing emergent vegetation, etc.) are influential factors. While the potential exists for distinguishing SAV species, health, or density, such projects require narrow, well-selected spectral bands from the imagery as well as very
detailed and robust ground-truthing.
Clark, C. D., H. Ripley, E. Green, A. Edwards, and P. Mumby.
measurement of tropical coastal environments with hyperspectral and high spatial
resolution data. International Journal of Remote Sensing. 18 (2): 237-
Fyfe, S. K. 2003. Spatial and temporal variation in spectral reflectance: Are seagrass species
spectrally distinct? Limnology and Oceanography. 48 (1, part 2) pp. 464-479.
Spectral processing methodology development for airborne imagery of
submerged aquatic vegetation in the Lower St. Johns River. Final Report to the St. Johns River Water Management District, Palatka, FL. Lewis, M.M. 2003. Hyperspectral discrimination of vegetation – what is possible? Proceedings of the EPA Spectral Remote Sensing of Vegetation Conference (pp. 148-151). Las Vegas, NV.
Session 7A – Page 3